Can your forecast take a punch? Unanticipated events like hurricanes and work stoppages can wreak havoc on a manufacturing forecast. Rich Weissman looks at how some companies stay on their feet June, 2006 - The Manufacturer US By: Rich Weissman

Jerry, the production manager at a New Jersey glass factory, heard the delighted squeals of his two school aged children when they realized that today's snowstorm had cancelled school. As he looked at the morning news, he saw that the storm had severely impacted the eastern United States, closing schools, airports, stores and even some manufacturing plants. The stranded truckers being interviewed on the television convinced Jerry that inbound and outbound logistics would also be impacted. The unplowed streets reminded him that his employees would have a hard time getting to work. Just as his kids were begging Jerry to go sledding, his thoughts turned to his manufacturing plan that was quickly deteriorating.

One major announcement from a competitor, one natural disaster, or even an event like 9/11 can impact, or even destroy, a forecast. Or, is it simply challenged? Some remarkable applications exist from enterprise software vendors and business intelligence suppliers, but are they enough to create forecasts that can 'take a punch' while still meeting operational and customer requirements?

Inaccurate demand forecasting causes great pain throughout the supply chain, causing manufacturing inefficiencies, misallocation of labor and materials, schedule fluctuations, unbalanced inventory, long lead times, and ultimately unhappy customers. Demand management techniques use forecasts to control, synchronize, and balance the supply and demand of a firm's products or services, including aggregate planning, master production scheduling, capacity planning and strategic product decisions. Demand management further allows for the closer coordination of the needs of the supply chain and customer requirements, resulting in increased efficiencies and process improvements.*

Forecasting and demand management impact distribution, especially in the service management business. Wauconda Illinois based supply chain management company Fidelitone Logistics provides distribution services for a diverse range of companies that include the nation's largest provider of home services, DeWalt and Black & Decker. "We are responsible to our clients to get their sales and service parts to their customers and that takes a lot of planning and forecasting," says Fidelitone Logistics' executive vice president Tom Giovingo. "We work closely with our suppliers to determine optimum inventory levels so we can meet service requirements."

Fidelitone Logistics uses 24-36 months of historical data to predict the future, but according to Giovingo there is no magic formula. "We need to look at a lot of factors that may impact our forecasting and ultimately our ability to service customers," he says. "Our suppliers are often launching major product promotions resulting in higher sales. It is imperative that we know about these promotions and plan accordingly."

Still, there is unpredictable usage that Fidelitone Logistics covers with safety stock, in its warehouses and at the supplier's site as well. "We keep one to two weeks of safety stock on hand for certain critical items in order to offset forecasting errors or an unanticipated increase in demand," says Giovingo. "We need to plan for snow storms and other events that impact our business, and there are some items where we have to attain a 'never out' status."

Forecasts are inherently an educated guess. Yet a poor forecast can have a damaging effect on manufacturing operations and customer satisfaction. An increase in the use of forecasting and demand management software applications, in addition to business intelligence, will best manage those forecast busting events. Those still using seat of the pants forecasting techniques may ultimately be down for the count.*

*Excerpts from the June 2006 article.